1 / 29

Understanding Network Failures in Data Centers: Measurement, Analysis and Implications

SIGCOMM 2011 Toronto, ON Aug. 18, 2011. Understanding Network Failures in Data Centers: Measurement, Analysis and Implications. Motivation. Motivation. $5,600 per minute. We need to understand failures to prevent and mitigate them!. Overview.

Download Presentation

Understanding Network Failures in Data Centers: Measurement, Analysis and Implications

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. SIGCOMM 2011 Toronto, ON Aug. 18, 2011 Understanding Network Failures in Data Centers: Measurement, Analysis and Implications

  2. Motivation

  3. Motivation $5,600 per minute We need to understand failures to prevent and mitigate them!

  4. Overview Our goal: Improve reliability by understanding network failures • Failure characterization • Most failure prone components • Understanding root cause • What is the impactof failure? • Is redundancyeffective? Our contribution: First large-scale empirical study of network failures across multiple DCs • Methodology to extract failures from noisy data sources. • Correlate events with network traffic to estimate impact • Analyzing implications for future data center networks

  5. Road Map Motivation Background & Methodology Results • Characterizing failures • Do current network redundancy strategies help? Conclusions

  6. Data center networks overview Internet Access routers/network “core” fabric Load balancers Aggregation “Agg” switch Top of Rack (ToR) switch Servers

  7. Data center networks overview Which components are most failure prone? Internet How effective is redundancy? ? What is the impact of failure? What causes failures?

  8. Failure event information flow • Failure is logged in numerous data sources Ticket ID: 34 LINK DOWN! LINK DOWN! LINK DOWN! Syslog, SNMP traps/polling Troubleshooting Tickets 5 min traffic averages on links Network event logs Network traffic logs Diary entries, root cause Troubleshooting

  9. Data summary • One year of event logs from Oct. 2009-Sept. 2010 • Network event logs and troubleshooting tickets • Network event logs are a combination of Syslog, SNMP traps and polling • Caveat: may miss some events e.g., UDP, correlated faults • Filtered by operators to actionable events • … still many warnings from various software daemons running Key challenge: How to extract failures of interest?

  10. Extracting failures from event logs • Defining failures • Device failure: device is no longer forwarding traffic. • Link failure: connection between two interfaces is down. Detected by monitoring interface state. • Dealing with inconsistent data: • Devices: • Correlate with link failures • Links: • Reconstruct state from logged messages • Correlate with network traffic to determine impact Network event logs

  11. Reconstructing device state • Devices may send spurious DOWN messages • Verify at least one link on device fails within five minutes • Conservative to account for message loss (correlated failures) LINK DOWN! DEVICE DOWN! Aggregation switch 1 Top-of-rack switch Aggregation switch 2 LINK DOWN! This sanity check reduces device failures by 10x

  12. Reconstructing link state • Inconsistencies in link failure events • Note: our logs bind each link down to the time it is resolved LINK UP! LINK DOWN! UP Link state DOWN time What we expect

  13. Reconstructing link state • Inconsistencies in link failure events • Note: our logs bind each link down to the time it is resolved LINK DOWN 2! LINK UP 2! LINK UP 1! LINK DOWN 1! UP ? ? Link state DOWN time 1. Take the earliest of the down times 2. Take the earliest of the up times How to deal with discrepancies? What we sometimes see.

  14. Identifying failures with impact Correlate link failures with network traffic • Summary of impact: • 28.6% of failures impact network traffic • 41.2% of failures were on links carrying no traffic • E.g., scheduled maintenance activities • Caveat:Impact is only on network traffic not necessarily applications! • Redundancy: Network, compute, storage mask outages Only consider events where traffic decreases BEFORE AFTER time DURING LINK DOWN LINK UP Network traffic logs

  15. Road Map Motivation Background & Methodology Results • Characterizing failures • Distribution of failures over measurement period. • Which components fail most? • How long do failures take to mitigate? • Do current network redundancy strategies help? Conclusions

  16. Visualization of failure panorama: Sep’09 to Sep’10 All Failures 46K Widespread failures Long lived failures. Link Y had failure on day X.

  17. Visualization of failure panorama: Sep’09 to Sep’10 Failures with Impact 28% All Failures 46K Component failure: link failures on multiple ports Load balancer update (multiple data centers)

  18. Which devices cause most failures? Internet ?

  19. Which devices cause most failures? Top of rack switches have few failures… (annual prob. of failure <5%) …but a lot of downtime! Load Balancer 3 Load Balancer 2 Load Balancer 1 Top of Rack 1 Top of Rack 2 Aggregation Switch Device Type Load balancer 1: very little downtime relative to number of failures.

  20. How long do failures take to resolve? Internet

  21. How long do failures take to resolve? Load balancer 1: short-lived transient faults Median time to repair: 4mins Load Balancer 1 Load Balancer 2 Top of Rack 1 Load Balancer 3 Top of Rack 2 Aggregation Switch Overall Correlated failures on ToRs connected to the same Aggs. Median time to repair: 5 minutes Mean: 2.7 hours Median time to repair: ToR-1: 3.6 hrs ToR-2: 22 min

  22. Summary • Data center networks are highly reliable • Majority of components have four 9’s of reliability • Low-cost top of rack switches have highest reliability • <5% probability of failure • …but most downtime • Because they are lower priority component • Load balancers experience many short lived faults • Root cause: software bugs, configuration errors and hardware faults • Software and hardware faults dominate failures • …but hardware faults contribute most downtime

  23. Road Map Motivation Background & Methodology Results • Characterizing failures • Do current network redundancy strategies help? Conclusions

  24. Is redundancy effective in reducing impact? Redundant devices/links to mask failures • This is expensive! (management overhead + $$$) Internet Goal: Reroute traffic along available paths X How effective is this in practice?

  25. Measuring the effectiveness of redundancy Idea: compare traffic before and during failure Measure traffic on links: • Before failure • During failure • Compute “normalized traffic” ratio: Compare normalized traffic over redundancy groups to normalized traffic on the link that failed Acc. router (backup) Acc. router (primary) X Agg. switch (primary) Agg. switch (backup)

  26. Is redundancy effective in reducing impact? Internet Less impact lower in the topology … but redundancy masks it Redundancy is least effective for AggS and AccR Core link failures have most impact… Overall increase of 40% in terms of traffic due to redundancy

  27. Road Map Motivation Background & Methodology Results • Characterizing failures • Do current network redundancy strategies help? Conclusions

  28. Conclusions • Goal: Understand failures in data center networks • Empirical study of data center failures • Key observations: • Data center networks have high reliability • Low-cost switches exhibit high reliability • Load balancers are subject to transient faults • Failures may lead to loss of small packets • Future directions: • Study application level failures and their causes • Further study of redundancy effectiveness

  29. Thanks! Contact: phillipa@cs.toronto.edu Project page: http://research.microsoft.com/~navendu/netwiser

More Related